Memories are made, at least in part, from changes in the distribution of synaptic AMPA-type glutamate receptors. In Alzheimer disease, these same receptors are subject to removal when amyloid-β (Aβ) builds up, leading to synapse loss (see ARF related news story). Now, a new study shows that mutations that reduce the activity of the GluR3 subunit of the AMPA receptor cause a different sort of cognitive problem, namely mental retardation. The work, from Tao Wang and colleagues at the Johns Hopkins University School of Medicine, Baltimore, Maryland, along with an international cast of collaborators, further strengthens the association of AMPA receptor loss and cognitive impairment. The paper appears in last week’s PNAS online edition.

The same journal carries a report linking a novel gene to neurodegeneration, through a pathway that is familiar to AD researchers. The pathway is endosomal trafficking and the gene is Vac14, a regulator of the signaling lipid phosphatidylinositol 3,5-bisphosphate (PI(3,5)P2). Lois Weisman and colleagues at the University of Michigan, Ann Arbor, show that Vac14 knockout mice develop large vacuoles in neurons, a result of defects in late endosomal transport. This is the same cellular domain where the sorting receptor SORL1 regulates APP processing (see ARF related news story). Polymorphisms in the SORL1 gene raise the risk of late-onset AD (see ARF related news story).

Although Vac14 is expressed in all cells, the knockout seems to affect neurons selectively, suggesting that these cells have a special weakness when it comes to deficits in endosomal function.

As described in the first report, the AMPA receptor channel mutations were discovered during a search for genes involved in X-linked metal retardation. Using cDNA microarray screening, first author Ye Wu identified one man with moderate mental retardation who showed a 10-fold reduction in expression of GRIA3, the gene encoding the ionotropic glutamate receptor subunit GluR3. Even though the investigators could not identify a mutation responsible for the low expression in that person, the discovery stimulated them to sequence the same gene from 400 additional unrelated men with X-linked mental retardation.

Doing this, they found one deletion and four mutations in GluR3, all missense mutations in highly conserved residues in functionally important areas of the protein. By expressing each variant in cultured cells, the investigators found that one mutation caused far lower expression of the Glu receptor, while the others encoded receptors with altered channel function. In whole cell recording experiments, two of these carried minimal or no current, and one displayed altered deactivation kinetics. When incorporated into heteromers with the GluR2 subunit, the mutant R3 proteins produced altered channel kinetics compared to wild-type R2/R3 receptors, showing that the mutations cause measurable changes in AMPA receptor function.

Patients with the various GluR3 mutations all showed similar clinical profiles, including moderate mental retardation, and muscular abnormalities. Thus, the authors conclude, significantly reduced iGluR3 channel function due to mutations in critical domains of the subunit is associated with a distinct neurological phenotype in humans. Whether the problem comes from failure of synaptic potentiation in the mature brain, or results from altered development is not clear, but further studies using mouse models may shed light on this question.

In the second report, lead author Yanling Zhang produced Vac14 knockout mice in order to investigate the physiological roles of PI(3,5)P2. Vac14, together with its partner protein Fig4, regulates the activity of the Fab1 kinase, which produces PI(3,5)P2 from PI3P. In the knockouts, the level of PI(3,5)P2 and PI5P were reduced by half compared to wild-type, while PI3P increased. Mice with Vac14 knockout displayed large vacuoles in neurons in parts of the brain and peripheral nervous system, and neuronal apoptosis. Some parts of the brain, notably the hippocampus and cortex, were less affected. This phenotype resembles that of the Fig4 knockout mouse, which further supports the idea that changes in PI(3,5)P2 are causing the effects. In addition, in humans, heterozygotic Fab1 mutations cause a mild disease where vacuoles appear in the cornea, again pointing to the importance of PI(3,5)P2 in vacuole formation.

Interestingly, all other organs in the mice appeared normal, despite the fact that Vac14 is present in all tissues. However, although Vac14-lacking fibroblasts appeared normal in vivo, they formed vacuoles in culture, and so the investigators used these cells to investigate the phenotype further. The vacuoles were derived from macropinocytotic vesicles, they found, and carried markers of late endosomes and lysosomes, suggesting that the vacuoles result from swelling of these organelles. The researchers also observed defects in trafficking of proteins from the trans-Golgi network to endosomes. This is the same pathway in which SORL1 and another Alzheimer-related protein, Vps35 function (Small et al., 2005).

Could PI(3,5)P2 have a role to play in AD? The idea is not far-fetched—other phosphoinositides have been linked to AD, most notably the 4,5 phosphates, where presenilin mutations linked to AD have been shown to cause an imbalance in PI(4,5)P which affects calcium channel activity in neurons (see ARF related news story).—Pat McCaffrey


  1. The intriguing study by Wu et al. provides a clear link between AMPA receptors, the major neurotransmitter receptor for rapid excitatory transmission in the brain, and specific genetic forms of mental retardation. The type of AMPA receptor examined, termed GluR3, is encoded by a gene on the X chromosome, and the authors found several different mutations in GluR3 associated with mild cognitive impairment which likewise influence receptor levels or electrophysiological properties. Since regulation of AMPA receptors contributes to synaptic plasticity and diverse kinds of behavioral learning, this connection is a logical one. It points to GluR3 as an important receptor for learning and memory.

    It is also interesting to note that disorders of cognitive impairment, including Alzheimer disease, have been recently linked to altered function of AMPA receptors at glutamatergic synapses, particularly in the hippocampus. It remains to be determined whether the genetic forms of cognitive impairment linked to mutations in GluR3 represent similar changes in synapse or circuit function as those attributed to Aβ exposure during AD. But clearly, further understanding the function and regulation of AMPA receptors holds promise for revealing the molecular basis for cognitive impairment and mental retardation, and thus for devising new therapeutic strategies.

    View all comments by Michael Ehlers

Make a Comment

To make a comment you must login or register.


News Citations

  1. AMPA Receptors: Going, Going, Gone in Aβ-exposed Synapses, PSD95 Knockouts
  2. Sorting Out SorLA—What Role in APP Processing, AD?
  3. SORLA Soars—Large Study Links Gene to Late-onset AD
  4. Beyond γ-Secretase: FAD Mutations Affect Calcium Channel via Lipid Messenger

Paper Citations

  1. . Model-guided microarray implicates the retromer complex in Alzheimer's disease. Ann Neurol. 2005 Dec;58(6):909-19. PubMed.

Further Reading

Primary Papers

  1. . Mutations in ionotropic AMPA receptor 3 alter channel properties and are associated with moderate cognitive impairment in humans. Proc Natl Acad Sci U S A. 2007 Nov 13;104(46):18163-8. PubMed.
  2. . Loss of Vac14, a regulator of the signaling lipid phosphatidylinositol 3,5-bisphosphate, results in neurodegeneration in mice. Proc Natl Acad Sci U S A. 2007 Oct 30;104(44):17518-23. PubMed.